US, China to discuss AI safety risks, cooperation on monitoring AI-directed cyberattacks

The US wants to discuss cooperation on monitoring AI-directed cyberattacks and has floated a proposal to ask US and Chinese AI labs to "police themselves" and share information.